7 papers
Recursive Scaling in Masked Diffusion Models
Alba Carballo-Castro, Julianna Piskorz, Paulius Rauba +2
Masked diffusion models (MDMs) have recently emerged as a promising paradigm for sequence generation. Scaling MDMs is conventionally achieved by increasing the parameter count or t…
Masks Can Be Distracting: On Context Comprehension in Diffusion Language Models
Julianna Piskorz, Cristina Pinneri, Alvaro Correia +3
Masked Diffusion Language Models (MDLMs) have recently emerged as a promising alternative to Autoregressive Language Models (ARLMs), leveraging a denoising objective that, in princ…
A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring
Usman Anwar, Julianna Piskorz, David D. Baek +6
Large language models are beginning to show steganographic capabilities. Such capabilities could allow misaligned models to evade oversight mechanisms. Yet principled methods to de…
Eliciting Numerical Predictive Distributions of LLMs Without Autoregression
Julianna Piskorz, Katarzyna Kobalczyk, Mihaela van der Schaar
Large Language Models (LLMs) have recently been successfully applied to regression tasks -- such as time series forecasting and tabular prediction -- by leveraging their in-context…
Improving the Generation and Evaluation of Synthetic Data for Downstream Medical Causal Inference
Harry Amad, Zhaozhi Qian, Dennis Frauen +3
Causal inference is essential for developing and evaluating medical interventions, yet real-world medical datasets are often difficult to access due to regulatory barriers. This ma…
Beyond the ATE: Interpretable Modelling of Treatment Effects over Dose and Time
Julianna Piskorz, Krzysztof Kacprzyk, Harry Amad +1
The Average Treatment Effect (ATE) is a foundational metric in causal inference, widely used to assess intervention efficacy in randomized controlled trials (RCTs). However, in man…